MRI Sign May Distinguish Hydrocephalus from Alzheimer's Disease
ABSTRACT & COMMENTARY
Source: Holodny A, et al. MR differential diagnosis of normal-pressure hydrocephalus and Alzheimer's disease: Significance of perihippocampal features. AJNR Am J Neuroradiol 1998;19:813-819.
Normal pressure hydrocephalus (nph) is one of the few causes of dementia that can sometimes be reversed by surgery. Suitable surgical candidates are usually identified by characteristic neuroimaging findings in combination with the clinical triad of dementia, incontinence, and gait disturbance. Unfortunately, not every elderly individual with enlarged ventricles and these symptoms responds favorably to shunting. Misdiagnosis can occur because cerebral atrophy and similar signs can be associated with disorders such as Alzheimer's Disease (AD). Holodny and colleagues investigated whether enlargement of the perihippocampal fissures (PHF) on MRI is useful for distinguishing NPH from AD. They focused on this region because hippocampal atrophy is an early and almost ubiquitous finding in AD, and it is usually associated with enlargement of the PHF.
The study involved 17 patients with shunt-responsive NPH and an equal number of age and cognition-matched AD patients. All participants had a Global Deterioration Scale score of 4-5, indicating mild dementia. The mean age of the NPH patients was 72 ± 6 years. All had been documented to have at least temporary improvement in their symptoms after shunting. The AD patients were chosen by the additional criteria of having a degree of ventricular enlargement on MRI comparable to that seen in the NPH cases. MRI images were obtained on 0.5 T and 1.5 T scanners, using T1 and T2 weighted sequences. The imaging plane was between 5 and 8 mm in thickness, and was, in most cases, oriented parallel to the long axis of the hippocampus. Subjective ratings of ventricular and PHF size were made by suitably trained neuroradiologists and their impressions were correlated with quantitative measurements of the same regions obtained by computer-assisted volumetry.
The investigators found that enlargement of the PHF distinguished AD from NPH in most cases. PHF volume averaged 1503 ± 720 cubic millimeters in AD and 423 ± 179 in NPH. If moderate-to-severe enlargement of PHF was used as the criteria for AD, PHF measurement had a positive predictive value of 86% in distinguishing NPH from AD. In contrast, measurement of total volume of the lateral ventricles or the volume of the third ventricle yielded predictive values of 79%, while the size of the temporal horn of the lateral ventricle did not significantly differentiate NPH from AD. Holodny and colleagues stressed that ratings by visual inspection were nearly as good as quantitative MRI volumetry in identifying significant PHF enlargement. They conclude that PHF volume may be a sensitive and specific neuroradiographic indicator, useful for distinguishing NPH from AD.
COMMENTARY
A variety of neuroradiographic correlates of NPH have been reported, including disproportionate enlargement of the temporal horns of the lateral ventricles, accentuated flow voids in the third ventricle, excessive transependymal fluid resorption, and other less dependable changes. None of these signs singularly predict response to shunting or exclude the possibility that ventricular enlargement is due to AD. This new study exploits the early occurrence of hippocampal atrophy in AD as a possible basis for resolving this differential diagnostic dilemma. Since AD accounts for as much as 75% of dementia in the elderly, while NPH is documented in 1% or fewer cases, improved keys to diagnosis would be valuable.
A small cohort size, as well as a retrospective, non-randomized design, limits the value of this study. Additional problems include the sampling bias inherent in selecting AD patients according to their ventricular size and failure to include non-NPH, non-AD neurological controls. Also, the study failed to include patients who didn't respond to shunting, a serious omission. Also, all patients with mild AD and large ventricles may not be unresponsive to shunting, since they may concomitantly suffer from NPH. Alert concludes that signs such as PHF enlargement on MRI may improve the differential diagnosis of NPH vs. AD but cannot be safely acted upon without additional measures. -nrr
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content